Yale’s Robert Shiller and Penn finance professor Jeremy Siegel have long dueled over whether stocks are cheap or expensive, and Daniel Fisher, of Forbes, reviews the arguments in his recent post. Shiller devised CAPE by measuring the inflation-adjusted earnings per share for the S&P over the trailing 10 years, instead of just the most recent quarter or year, and compared that number to long-run averages since 1871. The results showed a strong tendency for CAPE to revert to the mean — meaning when it was high, stocks tended to underperform going forward, and when it was below the long-term mean, they tended to do better. Siegel famously presented research to the Federal Reserve in December 1996, warning that stocks were dangerously overvalued.

Shiller’s CAPE is now at a worrisome level, but Siegel makes a strong case that the “E” in Shiller’s CAPE has been seriously distorted. Siegel says the ratio should be adjusted downward to account for the fact that several developments in accounting – most importantly, rules requiring companies to mark their investments to market on a quarterly basis – have made earnings much more volatile than the average since 1871.

The concentration of losses in the financial sector skewed earnings down, making the 10-year CAPE look high, according to Siegel. Even with Siegel’s adjustments the picture for stocks isn’t that exciting, however. Shiller’s CAPE suggests stocks will return an inflation-adjusted 2.2% a year through 2025. Siegel’s adjustments bring CAPE to 15.5-17.3x, suggesting investors could earn 5.25% real return. That’s still well below the 8% pension funds use as a guide, but comfortably above the 10-year Treasury Inflation-Indexed Security rate of 0.25%.